Objective: To investigate the accuracy and interobserver variability of Quantitative videodensitometric assessment of Mitral Regurgitation (QMR) after mitral valve replacement in a preclinical ovine model. Background: Quantitative videodensitometric aortography using the time-density principle is a well-documented technique for assessment of aortic regurgitation following transcatheter aortic valve replacement, but angiographic assessment of mitral regurgitation remains at best semi-quantitative (Sellers classification) and operator dependent. Methods: Fourteen sheep underwent surgical mitral valve replacement and were followed-up up to 1 year. A tricuspid valve with endogenous tissue restoration technology - the Xeltis valve (Xeltis BV, Eindhoven, the Netherlands) was implanted in 9 animals and the Trifecta valve (St. Jude Medical, St. Paul, MN, USA) was implanted in 5 animals. At the end of the follow-up period, the animals underwent a cardiac catheterization to measure their cardiac output and transmitral gradients immediately followed by ventriculography for QMR, then the animals were euthanatized, and the valve and peri-valvular tissue were excised. A mock-circulation (HDT-500 Hydrodynamic Test System [BDC Laboratory, US]) was used to assess the ex-vivo regurgitation fraction at a cardiac output comparable to the measurement obtained in-vivo. The CAAS A-valve 2.0.2 (Pie Medical Imaging BV, Maastricht, the Netherlands) was used for assessment of QMR. The QMR values (ex-vivo or in-vivo) as well as interobserver agreements were quantified after one, two, three, and four heart cycles. Results: The in-vivo QMR had significant correlations with the ex-vivo regurgitation fraction in all the cardiac cycles (best correlation: after 3 beats R2=0.784; p=0.002). The interobserver validation analysis showed significant agreements for all individual cardiac cycle (best agreement: after 4 beats R2=0.907; p<0.001). Conclusions: Angiographic quantification of mitral regurgitation using the CAAS-A valve prototype software is feasible, accurate, and highly reproducible in a preclinical setting. Further studies are necessary to establish the clinical value of this software.